Corporate social responsibility (CSR) is institutionalised amongst multinational corporations (Bondy et al. 2012). Today, 93 per cent of CEOs surveyed by the United Nations Global Compact (UNGC) say that acting in environmentally sustainable, socially responsible ways is vital to their firms’ success (UNGC and Accenture 2010, 2013). In 2013, 93 % of the world’s largest 250 firms produced a sustainability report (KPMG International 2013), publicly commenting on their achievements and shortcomings related to social, environmental, labour and human rights behaviours. Surveys of applied examples demonstrate that CSR is being adopted and implemented by corporations even whilst an irrefutable business case remains lacking (e.g. Carroll and Shabana 2010; Cowe and Hopkins 2008). Indeed, many firms regularly invest considerable funds in CSR-related programmes, and communities increasingly expect them to do so (Harvey and Bice 2014). In the mining and extractives industry—from which this article takes its examples—it is common for the world’s leading miners to allocate 1 % of pre-tax profits to community investment programmes, a figure which can stretch into the tens of millions (Bice 2013). Progressive regulation also incorporates notions of CSR. Transparency measures inherent in the Dodd–Frank Act in the US (US Congress 2010), human rights protections in the Law of the Right to Prior Consultation of Indigenous and Tribal Peoples in Peru, and concerns for a "social licence to operate" in regulation for Australia’s coal seam gas industry (COAG Standing Council on Energy and Resources 2012) all speak to the acceptance of social responsibility as a vital component of contemporary business.

The widespread acceptance of CSR says as much about the connection between social issues and business success as it does about changing social norms and values concerning “the corporation”, and indeed, how the corporation itself is understood by those who lead it (Swanson 2008). Emergent perceptions, such as growing shareholder concerns for socially responsible behaviour, lend legitimacy to CSR from a firm perspective and go some way towards giving firms “permission” to adopt, invest in and perpetuate socially responsible practices (Kurtz 2008). The recent emergence of benefit corporations, particularly in the US, speaks to the importance of integrating CSR into “business as usual”, especially from shareholder and legal accountability perspectives (André 2012).

Yet the conceptual and methodological challenges for CSR scholarship remain considerable, despite its widespread acceptance by businesses, communities and governments. Substantial advances in the study of CSR are hindered primarily by a few critical and recurring factors. First, analysis of CSR remains complicated by the use of various names to describe extremely similar practices and concepts (Melé 2008). For example, terms such as “sustainable development”, “corporate citizenship” or “corporate social performance” (CSP) are frequent substitutes, blurring the boundaries of what constitutes the scholarly field of CSR (Crane et al. 2008). Secondly, even within identified theoretical taxonomies—wherein one might expect agreement—conceptual integration is lacking and terminological differences persist (Garriga and Melé 2004). Thirdly, there is a scholarly tendency to frame CSR enquiries around what is essentially the same question: how can corporate investment—in terms of money, time and resources—in activities or issues that do not contribute directly to the financial bottom line be rationalised (Margolis and Walsh 2003)? In other words, what evidence exists to support a business case for CSR? Fourthly, many studies remain quantitative and focused at the macro level (Lockett et al. 2006; Aguinis and Glavas 2012), limiting our understanding of CSR’s on-ground operations and implications. Indeed, Godfrey and Hatch (2007, p. 87) argued in this journal that more micro-level theorising and research is necessary if we are to “progress in our understanding of CSR”. Finally, CSR still grapples with accusations of “greenwashing” (Marquis and Toffel 2014) or conflation with public relations, concerns which raise questions about its motivations and ethics (Joyner and Payne 2002).

The persistence of these core challenges has been attributed to lack of a unifying theory, limited measurement tools and underdeveloped empirical methods (see, for example, Crane et al. 2008; Jamali 2007; Waddock and Graves 1997). For instance, business case-focused studies are usually quantitatively geared to demonstrate a link between CSP and corporate financial performance (CFP; e.g. Orlitzky et al. 2003; Orlitzky and Benjamin 2001; Salzmann et al. 2005; Kurucz et al. 2008; Carroll and Shabana 2010). Other studies with this focus explore the business-related drivers which stimulate firms’ interest in CSR, such as forestalling regulatory "red" or "green" tape (e.g. Vogel 2005, 2008; Gunningham et al. 2004). Business case approaches may also be more instrumental, viewing CSR as a means of generating competitive advantage, whether through shared value creation (Porter and Kramer 2011), tapping frequently overlooked components of the supply chain (Prahalad 2003) or appealing to socially concerned investors (Kurtz 2008). Here, CSR may also be positioned alongside more traditional risk management (e.g. Husted 2005) or its potential influence on share value or stock desirability in the market (e.g. Kurtz 2008; Bilbao-Terol et al. 2013). Ethical approaches, meanwhile, assert that firms must consider social and ethical factors other than profit motive in decision-making and in action (e.g. Bird and Velasquez 2006; Carroll 1998; Donaldson and Dunfee 1999; Jones et al. 2005; Joyner and Payne 2002; Snider et al. 2003), or even that social responsibilities must be accepted by firms as an ethical obligation, “above any other consideration” (Garriga and Melé 2004, p. 53). A subfield of studies within ethical approaches focuses on how firms might better measure and account for ethical behaviour, including through public reporting and independent analysis of information (e.g. Zadek 1998; Hess 2007; Gray et al. 1996; Richardson 2004; Roberts 2006). Stakeholder approaches are also common and position CSR as encompassing a firm’s obligations to individuals and groups beyond shareholders (e.g. Clarkson 1995; Cooper 2004; Donaldson and Preston 1995; Waddock and Smith 2000; Jamali 2007). This orientation focuses CSR activities primarily on “community norms and laws” and on the role which stakeholders’ material issues play in driving managerial decisions (Dunfee 2008, p. 354).

Whilst a thorough review of the most common taxonomies of CSR theories is beyond the scope of this article, it is useful to consider these approaches’ core traits to situate the discussion of new institutionalist perspectives that follows. The defining tenets of business case, ethical and stakeholder models—including their definitions and boundaries of CSR, usual motivations, typical modes of implementation, view of the corporation and theory of change (Tolbert and Zucker 2005)—are outlined in Table 1.

Table 1 Common theoretical approaches to CSR: defining traits
Full size table

Whilst each of these approaches has its merits and limitations, a long-held focus on the business case paradox raises especial concern that we risk losing sight of what effects the CSR policies and programmes already being implemented by numerous companies are having on the communities they seek to help, the organisations which implement them, and on the industries in which those companies operate. If CSR is indeed institutionalised, at least amongst multinational corporations—as evidence now suggests—and if a business case remains lacking, what might be a different way to understand its adoption and perpetuation?

This article maps innovative pathways for scholarly inquiry by contributing a unique, social mechanism-based framework of CSR to the mainstream CSR literature, whilst building on the growing work of new institutionalist scholars in the field (Brammer et al. 2012). It theorises CSR as a pattern of distinct but overlapping social mechanisms which may operate differently in different contexts, but in ways which are readily identifiable as CSR. This is unlike other, common approaches to CSR in the mainstream literature which tend to conceptualise it as either a values framework (e.g. Hemingway and Maclagan 2004; Siltaoja 2006; Joyner and Payne 2002) or a set of particular business outcomes (e.g. Vilanova et al. 2009; Porter and Kramer 2008, 2011). Whilst others have fruitfully applied new institutionalism to CSR, this article is distinctive amongst these studies through its introduction of a social mechanism-based framework for CSR and through its multi-level approach. Through its focus on social mechanisms, the article aims to dive deeper into the institution of CSR itself, exploring how CSR becomes institutionalised and complementing and extending prior new institutionalist thinking on the subject.

Three key questions are explored to advance this aim. First, how might new institutionalist studies of CSR complement or enrich other approaches and respond to the key challenges in CSR scholarship, outlined above? Secondly, taking a deeper look into CSR as an institution, what are the main social mechanisms which institutionalise it? Thirdly—drawing on illustrative examples from a 3.5 year study of CSR in the global mining industry—what insights can be gained from application of a social mechanism-based framework to CSR?

In addressing these questions, the article works in the spirit of Campbell’s bricolage, adopting a “second movement” new institutionalist approach which draws upon a variety of paradigms (e.g. organisational, economic, discursive institutionalism; Campbell 2004; Campbell and Pederson 2001; Schmidt 2008). It pursues “problem-driven work” which can open doors to more meaningful practical knowledge (Davis and Marquis 2005). Following Margolis and Walsh (2003, p. 296), the article’s examination of CSR via the social mechanisms through which it arises and perpetuates is pitched at “middle range theory” in order to address the “considerable gap in our descriptive and normative theories about the impact of companies on society.” Thus, the investigation supports a multi-level analysis at industry (organisational field), corporation (formal organisation) and community (informal organisation) levels (Scott 1998) which helps to reveal and explain normative and political considerations which other types of theory may not (Campbell 2004; Campbell and Pederson 2001).

In undertaking this exploration, however, the article seeks not to assert that new institutionalism is necessarily the best or only lens through which to understand CSR. Indeed, exclusive taxonomic classification of CSR could continue ad infinitum, but the central point remains that segregated theories do not necessarily make the concept of CSR more coherent. In practice, such categories may encourage investigation of CSR in a piecemeal manner; a scholar examining the business case may avoid exploration of moral imperatives because this is seen as sitting outside of her theoretical scope. Thus, whilst the social mechanism-based framework introduced in this article is situated within new institutionalism, it is positioned not to the exclusion of other perspectives but to augment and allow deeper understandings. For example, the business case remains important; without financially viable business, there are no corporations to be responsible. Ethics remain paramount. Instrumental concerns, including competitive advantage, will retain salience as long as we retain faith in the market. And for each of these perspectives, an exploration of the key social mechanisms that institutionalise CSR may offer new insights and emphasise different lines of inquiry.

The article proceeds by first reviewing new institutionalist contributions to the study of CSR. It discusses how new institutionalism can address major scholarly challenges for the field. It then theorises a social mechanism-based framework for CSR, drawing on examples from a multi-level, multi-year study of CSR in the global mining industry. Benefits and limitations of the conceptual framework are discussed. The article concludes with a summary of key findings and suggests ways in which further application of a social mechanism-based framework to CSR could progress different and fruitful avenues for scholarly inquiry.

New Institutionalism: Addressing Core Challenges for CSR Scholarship

There is a limited but growing body of research applying new institutionalism to the study of CSR (see, for example the 2012 special issue of Socio-Economic Review on ‘Corporate social responsibility and institutional theory: New perspectives on private governance’), and anecdotal evidence suggests it is gaining traction amongst early career researchers interested in CSR scholarship (Brammer et al. 2012). Returning to the first research question, new institutionalist studies can go some way towards responding to the key challenges facing CSR scholarship, noted earlier.

First, amongst new institutionalist studies, there is a general conceptual coherence and agreement around CSR’s definition wherein it is understood as an identifiable pattern of activities influenced by social norms concerning firms’ social, environmental and economic behaviours and impacts, with CSR activities distinguished from a firm’s other activities by their very focus on these concerns, whether implicit or explicit (Matten and Moon 2008). This facilitates identification of a body of literature, encouraging discussion and debate (Brammer et al. 2012). Importantly, CSR is more than rule conformity, as here, the corporation is a multi-layered ‘agentic’ actor influenced by social norms and pressures (Meyer 2010) capable of varying degrees of conformity (Dacin et al. 2002; Oliver 1991). New institutionalism, therefore, assists our understanding of how institutions bound behaviours (Deephouse and Suchman 2008; DiMaggio and Powell 1991; Meyer and Rowan 1991). It helps to unpack the influences of intrinsic conviction (Frederick 1994) versus external forces (Aguilera and Cuervo-Cazurra 2004; King et al. 2005). Here, CSR is motivated by a complex web of social mechanisms which operate in concert to institutionalise it (Hedström and Swedberg 1998). This may manifest in corporate policies, industry or transnational governance frameworks, or via stakeholder engagement and regulatory compliance. Seen as a social process, CSR contributes to firm longevity, relationships, reputation and social roles (see, Table 1).

Secondly, new institutionalist studies tend to be framed around questions different to business case concerns. For example, the majority of new institutionalist studies of CSR focus either on macro pressures or on providing evidence of institutionalisation (Bondy et al. 2012). In these studies, firms are positioned within a broader social context and act out of self-interest influenced by a responsibility to others (Meyer and Jepperson 2000). They demonstrate varying degrees of agency (Tempel and Walgenbach 2007), from the passive objects of isomorphic pressures (e.g. Aguilera et al. 2007; Campbell 2007; Gardberg and Fombrun 2006) to highly strategic actors (e.g. King et al. 2005; Terlaak 2007; Dacin et al. 2002).

New institutionalist scholars also investigate questions concerning the linkages between corporate action and legitimacy, contributing a consolidated focus on key mechanisms or “institutionalising pressures” (e.g. Reuf and Scott 1998; Aguilera and Cuervo-Cazurra 2004). These studies relate to broader insights concerning the institutional scaffold which supports CSR in the form of market mechanisms, business institutions, civil society and state or government instruments (e.g. Gilbert et al. 2011; Waddock 2008). Studies drawn from rational choice and historical institutionalist perspectives (Thelen 1999) are particularly well-positioned to investigate the pressures shaping CSR in terms of norms, and explore the proliferation of certified management standards, such as ISO14001 or ISO26000, the UNGC or International Council on Mining and Metals (ICMM) 10 Principles for Sustainable Development (e.g. Terlaak 2007; Dashwood 2012a, b; King et al. 2005).

Thirdly, new institutionalist studies address concerns about CSR’s motivations and ethics by theorising the factors driving CSR’s adoption and perpetuation. Many look to macro-level influencers, examining the ways in which CSR’s forms and foci are shaped by national-level institutions (Matten and Moon 2008) or “varieties of capitalism” (Kang and Moon 2012). Those which position institutions as public/private or centralised/decentralised often situate firm responses to non-financial considerations as part of strategic behaviours (Ostrom 2014; Ingram and Silverman 2002; King et al. 2005). Studies focused on these concerns demonstrate how voluntary CSR standards may influence firm behaviour, even where norms may not (Terlaak 2007; Weiss 2000; Bendor and Swistak 2001; Parker 2002) and introduce the importance of decoupling between “rationalised myths” (Meyer and Rowan 1991) and practice (Terlaak 2007).

Others approach differing motivations by seeking to explain divergent practices within and between these macro-level systems (Jamali and Neville 2011; Turkina et al. 2015). The role of culture is therefore salient, and new institutionalist approaches facilitate exploration of the nexus of culture, symbols, values and institutions (e.g. Witt and Gordon 2009; Chapple and Moon 2005; Caprar and Neville 2012; Williams and Aguilera 2008). Through their focus on the normative and rational processes through which CSR is shaped, these studies contribute improved understanding of ethical behaviour and values considerations influencing managerial decision-making (Caprar and Neville 2012; Arnold et al. 2007; Christie et al. 2003). In so doing, their findings suggest CSR has advanced beyond public relations or greenwashing, at least for the most part. These investigations also often look to institutionalising mechanisms, beyond isomorphism, which may play a critical role in divergent CSR. Some explain degrees of CSR adoption and conformance in relation to standards and accountability (Gilbert et al. 2011; King et al. 2005; Terlaak 2007). Others consider lack of conformity to institutional pressures, variability within degrees of conformity, or even creation of hybridised forms of CSR through “cross-vergence” (Jamali and Neville 2011; Turkina et al. 2015; Aguilera and Jackson 2003). Studies like these are especially influential to the social mechanisms-based framework developed here, as they indicate the limitations of the explanatory power of isomorphism, alone.

Related to this, studies like Campbell's (2006, 2007) and Jamali and Neville’s (2011) interrogate CSR’s operation at multiple societal and organisational levels. Jamali’s (2010) work also encourages exploration of policy-practice decoupling and the diffusion of CSR throughout organisations. Such studies consider the roles of actors within organisations, facilitating consideration of corporate decision-making practices, the influence of individuals’ ethical values (Brunsson 1990; Carol 2008; Swanson 2008) and the role of institutional entrepreneurship (see, the 2007 special issue of Organization Studies, 28:7, on “Entrepreneurship as embedded agency”).

New institutionalist studies of CSR, like all conceptual approaches, have their limitations. Even within new institutionalism, the definition of “institution” remains debated and depends upon researchers’ paradigms and perspectives (Zucker 1977; Powell and DiMaggio 1991a; Campbell 2004). Nevertheless, new institutionalism’s staying power (Scott 2004) and growing collaboration amongst particular paradigms facilitates strong engagement and debate (North 1990; Campbell 2004). Whilst many new institutionalist studies do remain focused at the macro level (Williams and Aguilera 2008), calls for more meso- and micro-level theorising have been made (Terlaak 2007) and multi-level studies are appearing (Bice 2013; Koos 2012; Barley 2008). And whilst applications of national business system or varieties of capitalism models (e.g. Kang and Moon 2012; Matten and Moon 2008) prove to be poorly equipped to accommodate or explain divergent CSR practices within and between developing countries, very recent research is turning to this challenge (Turkina et al. 2015). Related to this, although new institutionalism is regularly employed to interrogate the global diffusion and adoption of certain ideas and practices, including CSR, it has historically paid only limited attention to the translation of such ideas and practices into local contexts (Brammer et al. 2012). An emerging subfield of research into the role of culture and divergence in CSR practices is taking up this challenge (Caprar and Neville 2012; Jamali 2010; Jamali and Neville 2011). New institutionalism has these and other limitations. Yet it “offers a promising avenue of research” for CSR (Brammer et al. 2012, p. 22), to which this article contributes.

Framework Development: A Social Mechanism-Based Framework for CSR

The above review of new institutionalist contributions to CSR scholarship indicates that the approach has much to offer in terms of addressing core challenges. This section of the article responds to the identified limitation of many new institutionalist studies’ tendency to target the macro level. Returning to the second research question, it takes a deeper look into CSR as an institution, and asks: what are the main social mechanisms that institutionalise CSR? By identifying and unpacking these mechanisms, the article generates a conceptual framework which can guide multi-level investigations of why CSR is adopted, how it is perpetuated and the extent to which its application, in practice, is coupled with societal or corporate values and policies. First, however, it is worthwhile to very briefly clarify the understanding of institutions which underpins a social mechanism-based framework for CSR.

Although debate continues, it is now generally agreed that institutions are not epiphenomena constructed by actors—the cumulative result of individual choice—but are sociological phenomena, shaping and shaped by social context (Powell and DiMaggio 1991a). The way in which institution is defined and understood is strongly influenced by the paradigm or discipline from which it is being examined. Thus, institutions may be defined variously as: frameworks of rules; proscriptive actions; patterns in repetitive interactions; customs; governance structures; social arrangements which minimise transaction costs; sets of norms, rules or principles; or directly or indirectly agreed roles combined with conventions (Powell and DiMaggio 1991b). Importantly, throughout these various definitions, institutions are usually not organisations or vice versa. For purposes of this article, an institution is defined as:

A social order or pattern (Jepperson, 1991) which is embedded in cultural and historical frameworks (Meyer and Rowan, 1977), is shaped by and shapes cultural norms, but which is not necessarily a product of strategic equilibrium (Campbell, 2004) or conscious design (Powell and DiMaggio, 1991a).

This definition of institution acknowledges the social pressures inherent within them whilst also recognising that, as social phenomena, institutions are not intentionally shaped. For example, whilst agendas may develop around CSR, they are not crafted directly through the work of individual, agenda-setting agents (Campbell 2007). Instead, institutions emerge as social patterns moulded by cultural norms. Consequently, they comprise various social mechanisms, each of which contributes an essential working part to the amalgamated whole (Hedström and Swedberg 1998). As interacting phenomena, mechanisms’ ultimate effects are not inherent in any one part (Davis and Marquis 2005).

For an institution to be formed, it undergoes a process of institutionalisation. Although bordering on the tautological, it is important to differentiate these terms. An institution is the social pattern which develops as a result of disparate social mechanisms working in concert. Institutionalisation, following Jepperson (1991, p. 145), is, “The process through which such a social pattern has attained the status of institution”.

Social Mechanisms Institutionalising CSR: Cases from the Global Mining Industry

Research Design and Case Study Data

Adopting a "second movement" approach (Campbell and Pederson 2001), the following section employs a range of new institutionalist perspectives, especially from organisational institutionalism, to define and examine the core social mechanisms which create a pattern identifiable as CSR. Although it is certain that the number of social mechanisms contributing to CSR is greater than those detailed here, those which are examined have been selected carefully, and the reasoning behind each selection is presented.

The global mining industry provides a particularly strong subject for an investigation of CSR from a new institutionalist perspective. The industry is characterised by multinational corporations which have institutionalised CSR through: membership in certified management standards (e.g. ISO14001, ISO26000), signatory memberships (e.g. UNGC) and industry principles (e.g. ICMM Principles for Sustainable Development); transnational governance commitments to ethical behaviour (e.g. Extractives Industry Transparency Initiative); widespread public accountability activities [e.g. Global Reporting Initiative (GRI) sustainability reporting]; organisational restructuring to prioritise CSR concerns (e.g. sustainable development or community relations departments); professionalisation of CSR, especially within community relations roles (Kemp 2010); implementation of corporate-level policies and CSR performance indicators at corporate headquarter and local site operation levels, amongst adoption of other common CSR values and behaviours (Bice 2014; Harvey and Bice 2014).

The social mechanism-based framework for CSR was developed, applied and tested through a 3.5-year study of CSR in global mining, conducted between 2008 and 2012. The study’s research design involved a multi-method, multi-phase approach in which findings from preceding research phases contributed to instrument development and lines of questioning in subsequent phases, helping to consolidate and test key assumptions and findings in a dialogic manner (Flyvbjerg 2001). Phase one concentrated primarily on the organisational field, garnering industry-wide data and comprising a document analysis of industry policy documents and relevant, global CSR initiatives, such as the ICMM Principles, the UNGC and ISO26000. A content and discourse analysis of 5 major, multinational mining companies’ sustainability reports published between 2004 and 2008 (18 reports) was also completed,Footnote 1 with 404,957 words or about 1157 pages analysed. Phase two focused on the formal organisational level and explored CSR at corporate headquarters. Data gathered in this phase consisted of a questionnaire concerning 10 literature-identified CSR concerns central to the mining industry and in-depth interviews with 11 senior corporate headquarter representatives of 9 major multinational mining companies. Phase three focused on the informal organisational level and involved two in-depth case studies at multinational miners’ operation sites in Australia, comprising a total of 40 in-depth interviews, a community life questionnaire completed by 49 community members, site visits and observations. This research design facilitated exploration of CSR at industry, corporate and operation site levels through examination of key social mechanisms.

Content and document analyses, qualitative methods and case studies have been demonstrated as especially strong methods for investigating institutionalising processes and institutional change (Campbell 2004; Schneiberg and Clemens 2006). Although the findings of this case study are not wholly generalizable—nor are they intended to be—they do allow for “analytical generalisations” (Yin 2003) which respond to the third research question’s concern about generating avenues for future inquiry.

Why Social Mechanisms?

Social mechanisms allow for a deeper application of new institutionalism to CSR. They can improve plausibility through explanatory power (Martin 2003), thereby strengthening our understanding of CSR, its operation and impacts. Importantly for future research, social mechanisms may facilitate construction of middle-range CSR theories (Davis and Marquis 2005) which could “offer an intermediary level of analysis in-between pure description and story-telling, on the one hand, and universal social laws, on the other” (Hedström and Swedberg 1998, abstract). In unpacking the main social mechanisms which institutionalise CSR, this article constructs a conceptual framework, allowing future studies to explore more fruitfully the impact of CSR at industry, organisation and community levels. This framework can help reveal and explain normative and political considerations which other types of theory may not.

In the sections below, the social mechanisms of discourse, mimesis, normative learning and coercion are introduced and discussed (see, Fig. 1, for an overview of these mechanism’s definitions and examples). Illustrative examples from the above research study are provided, as a complete write-up of the project is not possible given the scope and space constraints. Throughout the conversation, it is important to remain mindful that the social mechanisms comprising CSR intersect and overlap. This often makes distinct mechanisms difficult to distinguish from one another in practice. For purposes of theory building, we must rely on interpretation and a somewhat artificial (but ultimately helpful) disaggregation of social mechanisms in order to facilitate analysis and discussion. It is also important to keep in mind that social mechanisms operate and are affected by broader socio-cultural contexts.

Fig. 1
figure 1

Social mechanisms institutionalising CSR: definitions and examples in global mining

Full size image

Discourse

The mechanism of discourse refers not only to words, but to the style in which they are delivered, why they are used, how they are understood, the structures in which they are deployed and the agents who use them (Schmidt 2008). Discourse plays a major role in the ways in which institutions are translated in different contexts (Campbell and Pederson 2001), and is therefore steeped within cultural, historical and substantive circumstances (Schmidt 2008). For CSR in global mining, this means that discourse is perhaps both the most visible mechanism through which it is institutionalised and that which also contributes to varied understandings of CSR, especially between nations and cultures (Hutchins et al. 2005; Jenkins 2004). Discourse also operates at multiple levels, with industry bodies enshrining principles of sustainable development, corporate headquarters committing to sustainability reporting and operation sites talking in terms of responsibility to communities (Bice 2013).

At the industry level, contemporary discourse constructs CSR as a readily identifiable institution in global mining, even despite policies and management approaches which may differ between companies (Bice 2012). This institutionalisation occurs largely through the conscious and unconscious proliferation of language such as “sustainable development” and “beyond compliance” (Hutchins et al. 2005). Across countries and operating environments, mining companies’ CSR obligations are consistently referred to in terms of “being a good corporate citizen” or holding a “social licence to operate” (Yakovleva 2005).

In recent years the discourse of a social licence to operate has been particularly important in institutionalising CSR in mining, and is prevalent at multiple levels within the industry (Bice 2014; Bice and Moffat 2014; Prno 2013). At the organisational field level, it is deployed by peak industry bodies, such as the ICMM, which states a concern for companies’ “securing and maintaining a social licence to operate” (ICMM 2010a, p. 96). At the organisational level, individual companies use the term to represent the importance they place on identifying and addressing their social impacts and consequent social changes (Joyce and Thomson 2000; Owen and Kemp 2013). The term is prevalent throughout major miners’ public CSR communications, especially sustainability reports (Parsons and Moffat 2014; Bice 2014). Echoing sentiments in all of the studied reports, for instance, leading global miner BHP Billiton opens a recent sustainability report by stating:

Our ability to operate globally is dependent upon gaining access to natural resources and maintaining our licence to operate. Sustainable development is core to our business strategy; we integrate health, safety, environmental, social and economic factors into our decision-making (BHP Billiton, 2012, p. 2).

This language positing CSR as central to community relations, mitigation of environmental damage and operational viability was common across all companies’ reports studied. Yet it is also a language steeped within mining’s central concerns about natural resource exploitation, the historical controls of industry regulation, and an ultimate concern with business viability. As such, discourse plays an important role in shaping how CSR is understood and communicated within and by the mining industry, inside and between different companies.

Discourse also signals the power structures inherent within institutions and their flexibility for institutional change (Schmidt 2008). The users, types and adaptability of language used about CSR, therefore, play a considerable role in the concept’s currency and relevance. Where CSR is unable to adapt linguistically to shifting stakeholder expectations and corporate developments, its lifespan will be limited. Moreover, discourse necessarily operates at various societal and organisational levels. From presenting broad socio-cultural implications for capitalist systems (Jones 1999) all the way down to informing or constraining individual identity (Foucault 1998), discourse is vital.

As with the adoption of CSR language by peak mining industry bodies, discourse spreads through processes of diffusion and translation. This was seen across all studied companies in the spread of CSR language from international commitments to corporate headquarter policies to the daily language of community relations staff at mine sites. This diffusion represented the dissemination of largely unchanged institutional principles or practices through a related group of actors (Campbell 2004), potentially creating hypernorms (Wood et al. 2006). Translation processes were visible where globalised institutions were made sense of by local, regional and state actors (Campbell 2004). For example, local mine managers regularly referred to the broad responsibilities of their mining operation to the towns in which they are based, expressing both an expectation for good corporate behaviour but also noting the importance of their own role in the CSR process. For one local mine manager interviewed, CSR boiled down to being, “all about ensuring the community is looked after and people get involved”, whilst another corporate manager of sustainable development explained, “So, it’s the licence to operate which is the key driving factor for us. So, having the communities where we operate value us and wanting us to be there is our whole modus operandi”.

The process of translation, therefore, means that whilst CSR may play out differently according to context; it will always be identifiable but altered to suit the particular actors or organisations in question (Campbell 2004). Importantly, however, diffusion and translation cannot support unlimited proliferation. Instead they are bound by the national and international structures, regulations or institutions that promote CSR discourse. Where such support structures exist, socially responsible practices are more likely to be adopted and embedded (Jones 1999). As CSR discourse is shaped and reshaped through this structuration (Giddens 1984), the institution itself is transformed. Such processes are visible through widespread changes in CSR discourse over time—e.g. from corporate philanthropy to CSP, stakeholder management to corporate global citizenship (Carroll 2008). Discursive adaptations like these play a substantive role in how CSR is interpreted, adopted and implemented by organisations in different environments at different times.

The “social licence” example in the mining industry is again illustrative here. This terminology has been deployed by global miners since the early 2000s (Joyce and Thomson 2000). Its widespread acceptance reveals an important, industry-wide shift from earlier language centred on “health, safety and environment” towards a pseudo-regulatory licensing language (Bice 2014). In adopting such language, the mining industry’s current conceptualisation of CSR reflects contemporary global trends towards voluntary regulation (Vogel 2008) and stakeholder expectations about companies’ responsibilities to communities (Prno and Slocombe 2012).

Discursive mechanisms also emphasise the normative and cognitive schema which shape a shared social reality (Scott 1991). Following Berger and Luckman’s (1967) seminal work on the sociology of knowledge, discourse plays a major role in the “socially constructed reality” which shapes organisations and institutions (Meyer and Rowan 1977). Indeed, discourse often serves as the first contact point with any institution, and this holds true for CSR. Studies like that of Snider et al. (2003, p. 184) explore CSR web-based discourses in order to provide a “gestalt of the ways in which the most successful firms globally describe their corporate social responsibility”.

The mining industry represents this well. All major global miners now produce sustainability reports and all of the studied companies devote portions of their web pages to “sustainability”, “corporate responsibility” or “communities”. Managers interviewed for the study noted how these changes have normalised CSR in their organisations and other studies show that these approaches are now expected by stakeholders as a component of how mining companies do business with communities (Kemp 2010). Whilst normative and cognitive schema around a discourse of CSR in global mining is clear—the language is widely adopted and organised around particular, shared concerns—the depth to which such discourses are conceived is questionable. One general manager for sustainable development interviewed, for example, surmised these doubts, stating, “You can go to any [mining company] website and have a look at the sustainability drop-down menus and you think, ‘What a fantastic company!’ But you get out on the ground and there’s just such a disparity”. Research also suggests that, although landmark industry initiatives, such as the Extractive Industries Review (Salim 2004a, b) and most mining companies’ sustainability reports emphasise the importance of a social licence to CSR, little is done to define the criteria by which such a licence is granted (Bice 2014). Even so, the discourse has become so powerful, it is now being leveraged by community members as a means of holding mining companies to account (Moffat and Zhang 2014).

Social Mechanisms Influencing Isomorphism: Mimesis, Normative Learning, Coercion

The adoption of a shared discourse at multiple levels in the global mining industry is tightly linked to isomorphic pressures institutionalising CSR. In particular, mechanisms of mimesis, normative learning and coercion contribute to isomorphism. That is, they result in similarities between organisations and their environments and between organisations and other organisations (DiMaggio and Powell 1991). This process results in institutions which are slow to change, not easily influenced by exogenous factors and which limit the behaviours considered appropriate for actors. As Powell and DiMaggio (1991a, p. 11) describe it, “Institutions do not just constrain options: they establish the very criteria by which people discover their preferences”.

Importantly, to understand the social mechanisms contributing to isomorphism, it is critical to acknowledge that not only must organisations appear rational individually, they must also coexist within a network or field, wherein structuration—the conditions through which social systems are maintained or altered (Giddens 1984)—creates and constrains opportunities for autonomy (Burt 1992) and legitimacy (Scott 1991). Thus, isomorphism is bound by the interactions of actors within a given social system and is largely influenced by those actors’ efforts to achieve or retain legitimacy.

In the case of CSR, isomorphic pressures driven by mimesis, normative learning and coercion are the basis for theorisations about global convergence of CSR policies and practices (Matten and Moon 2008; Campbell 2007; Jamali and Neville 2011). Whilst new institutionalists differ in their opinions of whether structuration lends itself to diversity (Scott 2004) or homogenisation (DiMaggio and Powell 1991), in the case of CSR in the global mining industry, the latter appears to hold. The social mechanisms found to contribute to this isomorphism are explored, below.

Mimesis

Mimesis—the modelling of actors on one another—is closely associated with uncertainty (DiMaggio and Powell 1991) and is therefore most influential during the early stages of CSR’s development. In the global mining industry, CSR grew largely from a critical mass of international initiatives (Dashwood 2004; Schiavi and Solomon 2007) which were themselves responding to increased global media and public attention to mining-related impacts (Deegan et al. 2002). It is helpful, therefore, to consider briefly the historic development of CSR in mining to understand the influence of mimetic processes on CSR’s emergence in the industry.

A series of unfortunate events, including leaching of toxic chemicals into the Papua New Guinean Fly River downstream of BHP Billiton’s Ok Tedi mine (Banks and Ballard 1997), concentrated global attention on major miners, created a crisis of confidence and demanded response. Canadian author Havina Dashwood pinpoints 1998’s Global Mining Initiative and subsequent ICMM, and the mining, minerals and sustainable development project as the originating source of CSR in mining (Dashwood 2004). Smith (2008), writing in a special edition of the Yale Human Rights and Development Law Journal on “Corporate social responsibility in the extractive industries”, tightly links CSR’s birth with the spread of globalisation in the late 1990s, describing it as a bi-product of corporations going multinational. In these instances, the mining industry responded to increased pressure for socially and environmentally responsible behaviour, with major miners representing the early adopters (Dashwood 2012a) and a growing cohort of middle miners modelling their CSR on other firms over time. Managers interviewed for the study suggested that these modelling processes, and the shift from major to middle miners’ adoption, are visible, for instance, in the now widespread production of GRI sustainability reports. In 2013, the GRI recorded 179 sustainability reports produced in the mining industry, representing giants like BHP Billiton to middle miners like Indonesia’s ANTAM (see, GRI 2014).

Importantly, mimetic processes at early stages of institutional development may stymie innovation because adopters look to entrepreneurs that they “perceive to be more legitimate or successful” (DiMaggio and Powell 1991, p. 70). Such mimetic processes, therefore, lead to convergence of values and practices, even across diverse international contexts, and closely link CSR to efforts to achieve legitimacy. Convergence is also an important factor shaping CSR at local mine operation sites. As one general manager of sustainable development interviewed for the study explained, convergence around CSR practices is now so common amongst mining companies operating in similar regions, it has become difficult for individual companies to stake claims to benefits of their efforts for communities. “[It’s challenging] if you’re working in an area where you’ve got lots of companies and lots of organisations and government and everyone’s pulling their weight to address an issue, and you can’t really determine whether your work is what’s caused the change!”.

Convergence is visible in the global mining industry where companies’ CSR has been concentrated largely into a catalogue of agreed priorities and activities. The widespread adoption of common CSR practices by mining companies includes: local employment; procurement and shared infrastructure programmes; apprenticeships and training; donations and other charitable contributions to civil society groups, NGOs and other organisations; direct funding or delivery of welfare programmes; construction of civic infrastructure (e.g. hospitals, health clinics, schools, recreation centres, sports fields and non-employee housing); establishment and funding of company-controlled trusts, funds or foundations (i.e. local development NGOs); and support for “outsourced” community development services, particularly those associated with national and international (supply side) organisations (Harvey and Bice 2014).

The extent to which mimetic processes contribute to isomorphism in CSR practices in the global mining industry, therefore, raises important questions about the status and effectiveness of CSR priorities and activities now and into the future. Emerging research is beginning to question the form and effectiveness of commonly deployed mining company approaches to CSR (Bice 2013; Harvey and Bice 2014). From an institutional perspective, the strong isomorphism visible in mining companies’ and mining industry approaches to CSR suggests that the institution has advanced beyond the innovation stage, wherein early adopters were motivated by an entrepreneurial desire to improve performance. If this is the case—and the findings of this study and complementary research around a social licence to operate and related concerns about credibility would certainly seem to suggest this (Prno 2013; Owen and Kemp 2013)—then contemporary CSR in mining is better understood as “normatively sanctioned”, further supporting the argument for its institutionalisation.

Normative Learning

Normative learning has also contributed substantially to CSR’s institutionalisation. Linked closely to professionalisation, normative learning encompasses efforts to “define conditions and methods of work” (DiMaggio and Powell 1991, p. 70), as well as related processes such as establishing industry member organisations. The development of CSR in the global mining industry reflects such normative learning well. At the formal organisational level, over several decades, mining companies’ CSR has transformed from concentration on human resources practices and relatively limited public disclosure (Guthrie and Parker 1989) to comprehensive environmental, social and governance programming backed by progressively professionalised staff.

At the organisational field level, the establishment of the ICMM in 2001 created a member group comprising the world’s leading mining and metal companies “to advance their commitment to sustainable development” (ICMM 2014). This and other mining industry associations consciously and unconsciously assert isomorphic pressures concerning expectations for “best practice” upon the organisational field through membership and signatory commitments to global CSR frameworks, such as the UNGC or OECD guidelines for multinational enterprises. Myriad other pressures systemically influence multinational mining companies towards homogeneity (Shapiro et al. 2007).

Other key components of this normalisation include establishment of expected educational credentials and completion of particular degrees or training programmes. This is followed by similar role offerings across organisations, allowing for flows of personnel within an organisational field. For CSR, such normative processes can be seen in the offerings of top business schools, such as Harvard and Wharton (Harvard University 2011; The Aspen Institute: Centre for Business Education 2010), which teach “corporate responsibility” within core leadership modules. Others, like the UK’s Cranfield School of Management, offer students the opportunity to specialise in “sustainable business” (Cranfield University School of Management 2011). Further academic resources are allocated to the study of business’ contemporary role within society in places like Boston College’s Center for Corporate Citizenship or the University of Queensland, Australia’s Centre for Social Responsibility in Mining. Graduates of these and similar programmes enter a job market in which leading mining companies now employ individuals to fill roles such as “manager, environmental and social responsibility”, “director, corporate social responsibility” or “environment, health and safety specialist” (see job listings on: BrighterPlanet.org, for example).

Coercion

Coercion contributes to isomorphism through “formal and informal pressures exerted on organizations by other organizations upon which they are dependent and by cultural expectations in the society within which organizations function” (DiMaggio and Powell 1991, p. 66). Coercion, therefore, often plays out via regulation. For purposes of our discussion, regulation is not necessarily formal or codified in law. Instead, coercive regulatory mechanisms include formal codes and laws and those voluntary guidelines, performance indicators and principles or frameworks which set standards for CSR performance.

In practice, “voluntary” regulation is perhaps more strongly associated with CSR than its more formal counterpart (Eigen 2007). Such regulation helps to define areas of concern; creates recognised minimum standards; supports a shared dialogue amongst organisations; suggests a particular global order; fosters accountability and competition through improved performance comparability; may seek to address deficient regulatory or governance standards (Aguilera and Cuervo-Cazurra 2004); may equally devolve or divert authority from the level of the state to the industry or organisation (Vogel 2008); and allows space for the establishment of standards which may be more industry specific and, by extension, more rigorous than generic, formal regulation (Brunsson and Jacobsson 2000; Campbell 2007).

Much research concerning the regulation of CSR has dealt primarily with the debate over whether regulatory codes should be requisite, as opposed to voluntary (e.g. Eigen 2007; Sethi 2003; Hess 2007; Sadler 2009). A new institutionalist perspective on the regulation of CSR suggests that, regardless of whether regulation is legislated by the state or introduced by private or non-profit organisations, its proliferation and uptake by corporations is what matters (Campbell 2007).

The mining industry represents this perspective well. As more and more companies choose to adopt particular principles or report against certain frameworks, interviewees reported that the peer pressure to use those frameworks increases. This diffuses values, discourse and expectations across the industry and increases coercive isomorphic pressures (Galaskiewicz 1991). Related to these findings, studies like that of Özen and Küskü (2009) suggest that, where industries are highly concentrated and competition is high—as in the global mining industry—voluntary regulation is more likely to be adopted as a means of legitimacy seeking.

The uptake of voluntary regulation of CSR in the mining industry is also likely linked to its sheer volume of proliferation. In his 2008 meta-analysis, David Vogel cited approximately 300 “global civil regulation” codes intended for the voluntary regulation of CSR. As one sustainable development manager interviewed for the study complained, “Over the last 10 years the industry generally, and [our company], have developed from doing the basic EIA (Environmental Impact Assessment) to socio-economic impact assessment, and all of the acronyms as well that we can have”. In practice, the most common global voluntary regulations concerning CSR include the GRI and the UNGC (GRI 2010; UNGC and Accenture 2010), in addition to formalised requirements for impact assessments. Frequently used together (GRI and UNGC 2006), these frameworks foster consensus around the key issues on which corporations seeking to act responsibly should focus. Following Power (1997), the GRI as an auditing tool can be viewed as an “institutionalised product”, which is used for “external legitimation”. Such regulated standards lend credibility to mining companies’ CSR efforts whilst also disseminating a global CSR discourse, linking and homogenising patterns of socially responsible behaviour through exertion of institutional pressure.

Coercion is also closely connected to societal expectations and the exercise of coercive power, as opposed to authority, as in regulation (Scott 1991). Such coercion via social pressures is evidenced in the historical adoption of CSR in global mining, discussed above. At the informal organisational level, the responses of companies to community pressures illustrate the role and agency of stakeholders in influencing the adoption and perpetuation of CSR (Delmas and Toffel 2004). For instance, community interest groups such as “No dirty gold” and “Lock the gate” act as civil society watchdogs. Non-governmental organisations, including Oxfam Australia’s mining campaign, Transparency International, Publish What You Pay and Save the Children, place the weight of their organisations behind social pressures for environmentally and socially responsible mining. Furthermore, publications such as Business for CSR, CSRWire, Ethical Corporation and Ethical Performance now regularly monitor companies’ social and environmental successes and failures. Intergovernmental agencies and international standards organisations such as the United Nations, World Bank, International Finance Corporation and International Organization for Standardization all play a major role in encouraging minimum benchmarks for businesses’ CSR behaviour through codes and standards. These aim to establish consistency of approach, to improve comparability between companies and industries, and to build credibility and legitimacy.

The coercive mechanisms that institutionalise CSR point to both causal factors (Campbell 2007) and isomorphic pressures. In particular, the voluntary frameworks and standards widely adopted by multinational miners, such as the GRI and ICMM principles (2010b), normalise and reinforce cognitive schema concerning the issues, practices and reporting indicative of socially responsible companies. Such cognition through regulation (even where it is not regulation by the state) creates a degree of stability, promoting isomorphism in which organisational agents look to other organisations’ agendas, interests and behaviours when making decisions about their own actions or choices (Fligstein 1991). Like the timeless “chicken or egg” question, however, it is impossible to determine with any certainty the extent to which frameworks like the GRI shape the focus of mining companies’ CSR activities, or the extent to which pre-existing CSR behaviours influence the frameworks themselves. Regardless of the direction of force, however, what is clear is that voluntary regulation in the form of reporting frameworks, societal pressures exercised by community and organisational stakeholders, and subsequent behaviours reinforce a cycle in the global mining industry in which CSR is reified, legitimised and institutionalised.

Conclusion: Opportunities and Limitations

This article has articulated common challenges for CSR scholarship and discussed recent contributions of new institutionalism to the field and its capacity to address core challenges. The article has then contributed a unique framework of the key social mechanisms institutionalising CSR: discourse, mimesis, normative learning and coercion. Drawing on illustrative examples from the global mining industry, the article demonstrates the ways in which these mechanisms have worked in concert to institutionalise CSR at multiple levels in this sector.

Revisiting the third research question, the social mechanism-based framework for CSR holds considerable opportunities for its study. Keeping with the mining industry examples presented above, future research applying the framework could progress understanding on a variety of emerging CSR-related issues. For example, studies might apply the framework to investigate further how the discourse of social licence to operate is creating new CSR motivators in a changing institutional environment. The framework could be deployed in comparative CSR studies to better consider the role of culture and values in influencing the form and foci of CSR, especially in frontier economies home to burgeoning resources sectors. The framework would support further exploration of CSR’s adoption and perpetuation at multiple levels, fostering investigation of the pressures and expectations of certain stakeholder groups at the community level and the extent to which those activities influence corporate headquarter or organisational field-level CSR policies and practice. Other studies might apply the framework to timely questions of regulatory efficiency, examining how discourse, isomorphic and regulatory mechanisms may influence institutionalisation of red or green tape cutting agendas, for instance. These are but a few of the scholarly opportunities latent in a social mechanism-based framework for CSR which could be applied well beyond the mining sector.

The discussion, however, is also limited by its application of the proposed framework to a single industry, although mining was carefully selected due to the centrality of CSR concerns to its operations and impacts. The utility and appropriateness of the social mechanisms-based framework for CSR could certainly be tested and strengthened through application to other sectors. In particular, this would facilitate testing of whether the particular social mechanisms selected into the framework are of similar meaningfulness for other sectors. Or, would other considerations, such as time or markets, be more helpful (Hedström and Swedberg 1998)?

The framework itself also has certain limitations. The permeability of boundaries between social mechanisms demands an ideal-type application and a suspension of the reality in which various mechanisms overlap. But, as the above discussion shows, such application is fruitful for analysis, as long as these caveats are made and ambiguities acknowledged. In the case of mining companies’ sustainability reporting, for instance, it is difficult if not impossible to ascertain whether mimetic (e.g. a critical mass of companies are now reporting) or coercive (e.g. demands of ICMM membership) processes are the primary driver for the diffusion of sustainability reporting across the industry. Although the framework—in its current form—may be unable to facilitate a prioritisation of various mechanisms’ operations, it demonstrates strongly that these mechanisms are at work and that they are important for a deeper understanding of CSR at multiple levels.

Finally, the framework also requires a general acceptance of the usefulness and relevance of new institutionalism to CSR. Although a strong case has been set out here—and advanced by many others—the contested field of scholarly CSR research demonstrates the challenges in asserting one particular approach over another. Other, plausible explanations could certainly be posited for certain of the ideas presented. For example, in relation to isomorphism, a scholar of organisational culture might assert that adoption of particular standards or practices is not coercive but is instead a means of forming a profitable organisational identity (e.g. Hatch and Schultz 2002) or for establishing corporate branding (Hatch and Schultz 2003). This is but one of myriad alternative interpretations. But as an integrated whole, the framework presents a unique tool for teasing out the complex interplay of different and sometimes competing pressures which shape CSR at multiple levels.